Understanding why men and women respond differently to the same cancer therapies has long frustrated oncologists. A large-scale proteogenomic analysis now offers a molecular explanation — not just at the level of gene expression, but at the protein level, where drug targets actually live and therapeutic vulnerabilities are decided.
Drawing on 1,590 tumor proteomes from eight cancer types within the Clinical Proteomic Tumor Analysis Consortium, the researchers mapped sex-differential protein abundance across tumors including lung adenocarcinoma, hepatocellular carcinoma, clear cell renal carcinoma, pancreatic adenocarcinoma, lung squamous cell carcinoma, and glioblastoma. The most striking disparity emerged in lung adenocarcinoma, where 901 genes showed sex-differential protein expression. In male tumors, those proteins clustered around the MYC and E2F oncogenic transcription factor pathways — both well-established drivers of aggressive cell proliferation. Female tumors, by contrast, showed enrichment in metabolic regulation and stress-response proteins. A portion of these protein-level differences were traceable to sex-differential copy number aberrations in the underlying tumor genome. Crucially, CRISPR-based gene dependency screening indicated these differentially abundant proteins also represent meaningful survival vulnerabilities — raising the prospect of sex-stratified therapeutic targeting.
This work matters for several reasons beyond its scope. Prior cancer sex-difference research has focused predominantly on the genome or transcriptome; the proteome is where biology is executed. Protein abundance does not always correlate with gene expression, so these findings add a distinct and actionable layer. The MYC pathway enrichment in male lung adenocarcinomas is particularly notable — MYC remains one of the hardest oncogenes to drug, and identifying that its targets are more proteomically prominent in male tumors could inform patient selection in emerging MYC-directed trials. The study's retrospective, observational design and reliance on publicly available datasets are legitimate limitations, and causal inference requires prospective validation. Still, with 1,590 proteomes across eight cancer types, this is among the most comprehensive sex-stratified cancer proteome analyses published. It signals that sex should be treated as a biological variable in every phase of cancer drug development — from target identification through clinical trial design.